The goal of this trial is to test the effect of brief physician advice on alcohol-related behaviors and health status of problem drinkers. Subjects will be selected from the practices of 60 primary care physicians located in southern Wisconsin. Approximately 320 patients from each practice with regularly scheduled appointments will complete an imbedded alcohol screening instrument which contains questions on smoking, weight loss, exercise, and alcohol use. The trial is focused on the treatment of adult male problem drinkers, defined as subjects who drink more than 20 drinks per week. Following a face-to-face interview of those subjects who score positive on the alcohol screening instrument, subjects will be randomly assigned to an experimental or a control group. The experimental group will be exposed to a standardized intervention protocol administered by their physician. A sample of 540 subjects is needed in each group to achieve an 80% chance of detecting a 10% difference in alcohol use. The protocol includes a diagnostic interview, a brief intensive counseling session, a self-help manual, a prescription to change their drinking habits and two follow-up visits. Fourteen outcome variables were selected based on the findings of completed clinical trials. These variables include: 1) number of weeks of sobriety; 2) number of episodes of binge drinking; 3) alcohol use in 7 days prior to six- and twelve-month follow-up telephone interviews; 4) days of hospitalization; 5) sick days; 6) injuries and accidents; 7) emergency room visits; 8) legal events; 9) global perception of health; 10) number of outpatient visits; 11) number of prescriptions; 12)self-reported health status; 13) family relationships; 14) number of chronic illness visits. The outcome variables will be assessed by a self-report telephone interview at six and twelve months, a telephone interview with a family member as twelve months, a review of medical records and a review of county and state legal records. Laboratory measures were not included for the following reasons: 1) poor psychometric properties of currently available tests in primary care populations of problem drinkers; 2) practical issues of recruiting subjects in a primary care setting; and 3) resource limitations. The analysis will be based on the development of a performance model of the outcome variables of interest. The proposed trial would be the first large scale randomized clinical trial conducted in the United States to test the effect of brief physician advice with problem dhnkers in a non-academic primary care setting.